1 / 32

Roberta Arnaldi INFN Torino

Challenges in quarkonium suppression studies at LHC. Roberta Arnaldi INFN Torino. Quarkonium production: Probing QCD at LHC, Wien, 17-21 April 2011. Quarkonium suppression in AA collisions.

verda
Download Presentation

Roberta Arnaldi INFN Torino

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Challenges in quarkonium suppression studies at LHC Roberta Arnaldi INFN Torino Quarkonium production: Probing QCD at LHC, Wien, 17-21 April 2011

  2. Quarkonium suppression in AA collisions Quarkonium suppression is considered since a long time as one of the most striking signatures for the QGP formation SPS RHIC LHC 17 GeV/c 200 GeV/c 2.76 TeV/c √s 1986 years 1990 ~2000 2010 Quarkonium study in AA collisions is already a 25 years long story. Many results were provided by SPS and RHIC experiments, but the picture is yet not clear! Can the LHC play a leading role in quarkonium suppression studies, providing an answer to the open questions?

  3. SPS heavy ion experiments A long heavy-ion program has been carried out at SPS and several experiments (NA38, NA50, NA60) were focused on charmonia study NA50 Several years of data taking (1996-2000) PbPb collisions @ 158 GeV Number of collected J/ 1995  ~ 50000 1996  ~ 190000 1998  ~ 40000 2000  ~ 100000 J/width ~ 100 MeV/c2

  4. How NA50 studied J/ suppression? J/ suppression studied comparing J/ yield to the Drell-Yan one Advantage: DY insensitive to the medium, efficiency and luminosity terms cancel out in the ratio 8 bins Drawback: the small Drell-Yan statistics limits the number of centrality bins Reference for the “anomalous” suppression is the J/behaviour in cold nuclear matter, extracted from pA data Results presented as a function of several centrality estimators Idea was to understand the variable driving the suppression

  5. One step further: the NA60 experiment NA60 Based on the NA50 experimental apparatus improved with a pixel vertex detector in the target region High quality data, thanks to the improved experimental apparatus • ~30000 J/ • J/ width ~70MeV • ~5% background+DY under the J/ Lighter colliding system: In-In @ 158GeV to get further insight in the J/ suppression comparing lighter and heavier systems at the same energy

  6. How NA60 studied J/ suppression? Not enough Drell-Yan statistics (~300 ev. M>3.2 GeV/c2) to study the centrality dependence of the J//DY ratio The J/ centrality dependence is directly compared to theoretical expectation based only on cold nuclear matter effects (Glauber calculation)

  7. SPS results Let’s compare NA50 and NA60 final results: To understand anomalous suppression, the reference determination is crucial  important step further with the reference based on NA60 pA data @ 158GeV, the same energy of the AA data In-In 158 GeV (NA60) Pb-Pb 158 GeV (NA50) Anomalous suppression visible for central PbPb collisions Agreement between PbPb and InIn in the common Npart region PbPb data not precise enough to clarify the details of the pattern! After correction for EKS98 shadowing

  8. J/ studies at RHIC PHENIX/STAR Similar strategy as the one adopted at SPS: AuAu @ s=200GeV CuCu @ s=200GeV  lighter system pp @s=200GeV  for reference dAu @s=200GeVto determine cold nuclear matter effects PHENIX J/e+e-|y|<0.35 & J/+- |y| [1.2,2.2] STAR J/e+e-|y|<1 Results based on a smaller statistics wrt SPS ~ 15000 J/ in the forward region ~ 1000 J/ at midrapidity arXiv:1103.6269 The J/ suppression is studied through the nuclear modification RAA or the RCP factors Recent comparison with dAu data, in order to account for cold nuclear matter effects in AuAu

  9. RHIC results arXiv:1103.6269 Stronger suppression at forward rapidity than at midrapidity No satisfactory theoretical description Suppression larger than CNM expectations Comprehensive understanding of the numerous CNM effects not yet available  quantitative estimate of hot matter effects still missing! In a similar way, as at SPS, CNM have been evaluated from dAuRCP suppression beyond CNM similar at y=0 and at y=1.7 breakup has a strong y dependence and it is larger at forward rapidity Recent results shown that CNM do not properly describe RdAu, making their extrapolation to AA more uncertain (J. Nagle, arXiv:1011.4534)

  10. Comparison SPS and RHIC:open questions Have SPS and RHIC results already provided a clear picture of J/behaviour in a hot matter? Comparison done also in terms of * Bjorken energy density  complicate issue, in particular comparing several experiments Results are shown as a function of the multiplicity of charged particles (~energy density, assuming SPS~RHIC)

  11. Theoretical interpretations Several theoretical models have been proposed to explain the similar suppression at SPS and RHIC: 1) Only J/ from ’ and c decays are suppressed at SPS and RHIC  same suppression is expected at SPS and RHIC  reasonable if Tdiss (J/) ~ 2Tc • 2) Also direct J/ are suppressed at RHIC but cc multiplicity high  J/ regeneration ( Ncc2) contributes to the J/ yield The 2 effects may balance: suppression similar to SPS Unfortunately data do not allow to clearly assess if recombination can play a role at RHIC J/ elliptic flow  J/ should inherit the heavy quark flow Recombination is measured in an indirect way J/ y distribution  should be narrower wrtpp J/ pT distribution  should be softer (<pT2>) wrtpp

  12. What have we learnt at SPS and RHIC? After the study of AA collisions at SPS and RHIC, several lessons have been learnt…even if the J/ picture in the hot matter is not yet clear The understanding of CNM effects is crucial to correctly quantify the hot matter effects • Complicate interplay between different effects, with strong kinematic/energy dependences •  not yet well defined • have to be evaluated from pA data at the same energy as AA • can pA data be used at all ? High precision data are required to precisely define the centrality dependence of the J/ yield  Well defined centrality pattern are needed to compare lighter/heavier systems or results as a function of different variables Deeper understanding comparing different experiments?  In principle yes…but different CNM effects, different systematics, not straightforward comparison in terms of energy density…

  13. T/TC J/(1S) c(1P) ’(2S) What should we expect at LHC? ...many questions still to be answered at LHC energy! • Role of the large charm quark multiplicity •  •  will J/ regeneration dominate the J/ picture? Regeneration? Further suppression? Role of other quarkonia states (in particular bottomonium) • still (almost) unexplored in HI collisions • (<100 (1S+2S+3S) in AuAu@200GeV – STAR)

  14. ALICE ATLAS CMS The HI program at LHC Three of the LHC experiments foresee to study quarkonium in AA collisions J/+- 2.5<y<4 J/e+e- |y|<0.9 pT coverage down to pT~0 ALICE pT>3GeV, ||<2.5  pT J/>6.5GeV/c ATLAS J/+- |y|<2.4 (separation between B and prompt J/) pT coverage depending on the y region CMS J/+- |y|<2.4 (separation between B and prompt J/)

  15. The 2010 PbPb LHC run The first LHC heavy ion run took place in November 2010 PbPb collisions at s=2.76 TeV Similar integrated luminosity for ALICE, ATLAS, CMS ~9 b-1 Peak luminosity increasing from 3x1023 to 3x1025 Hz/cm2

  16. 2011-2012 Heavy-ions LHC run First schedule discussed at the 2011 Chamonix workshop. Two years of data taking before the 2013 long shutdown 2011 It should be possible to obtain the following peak and integrated luminosity (presently foreseen 4weeks of data taking) L = 1-1.4 x 1026 cm-2 s-1 ~ factor 4-5 improvement wrt 2010 (due to more squeezed beams and a higher bunch number) 2012 Data taking scenario not yet clear! Two possibilities under study: • PbPb data taking to reach the goal of 1nb-1 data (if a luminosity increase can be foreseen) • p-Pb (Pb-p) data taking? (M. Ferro-Luzzi, J.M.Jowett, 2011 Chamonix workshop)

  17. First J/ invariant mass spectra Expected J/ width ~ 50-80 MeV/c2 S/B depends on centrality (~0.2 – ~3)

  18. Which J/ statistics has been collected? 2010 Let’s start from ATLAS published result (Phys.Lett B 697 (2011)294): pT>3GeV, ||<2.5  pT J/>6.5GeV/c Analyzed L~6.7 b-1 measured J/ ~ 600 CMS acceptance is similar, so let’s assume a similar number of J/ (both ATLAS and CMS should be able, as in pp, to reach a lower pT, in a restricted y range, but with a larger background contribution) ALICE has a different acceptance region, down to pT J/ =0 Focusing on the dimuon channel and taking into account that the J/ production cross section is higher at low pT , we can assume, roughly, a factor 2 more wrt ATLAS measured J/ ~ 1500

  19. …and in 2011? 2011 Let’s assume the same running efficiency, and the same beam qualityas 2010 The machine will deliver L~30-50 b-1 ATLAS/CMS (for pT J/ >6.5GeV): ~2000 – 3500 J/ ALICE (2.5<y<4): ~5000 – 8000 J/ What can be done with this J/ statistics?

  20. How long should LHC run to reach SPS statistics? Based on these numbers, which is the luminosity that LHC should collect in order to have the SPS J/ statistics? Let’s consider the NA50 and the ALICE case: NA50 (2000 data taking): 100000 J/ To collect this number of J/ in ALICE, we would need an integrated luminosity 100000 J/ L ~ 600 b-1 Hardly reachable unless a large luminosity increase will be provided by the machine with the design luminosity (L= 5x1026 cm-2s-1) this would require 106s assuming a 12% running efficiency, this means ~ 3 months (1month/year devoted to nuclear collisions at LHC)

  21. How many centrality bins can be done? Let’s start from the ATLAS paper (but similar evaluations hold also for ALICE/CMS): 2010 NJ/~600  4 centrality bins (0-80%) • ~90 – 200 J/ in each bin • Statistical error ~10% • System. uncertainty ~8% Total Error ~13% What should we expect in 2011? Let’s consider the most optimistic case: NJ/=8000  it’s possible to increase the number of bins 2011 Smaller statistical error, but we will be dominated by the systematic uncertainties  ~10% total error? …neglecting in these estimates any systematic uncertainty related to the unknown J/ polarization

  22. What can we study? The easiest approach is the study of the RCP ratio, to study the relative centrality dependence of the pattern: Atlas, Phys.Lett B697(2011)294-312 = Non centrality-dependent systematic uncertainties cancel out To quantify the PbPb modifications, wrtpp, the RAA has to be evaluated = More complicate result because it strongly depends on: • systematic uncertainties evaluation • J/ production cross section in pp collisions at 2.76 TeV

  23. The reference: 2.76 TeVpp data The correctly quantify the J/behaviour in AA, it is crucial to have a well defined reference J/ in proton-proton at the same centre of mass energy as PbPb Two possibilities: In March 2011 LHC has provided pp collisions at 2.76TeV  advantage: already at the same PbPb energy  drawback: only few days of data taking. Enough J/? Evaluate J/ at 2.76 TeV, relying on the measurement at 7 TeV (high statistics) and rescaling it via FONLL and CEM calculations (F. Bossu’ et al. arXiv:1103.2394) Systematic uncertainty on the extrapolated value is ~15%

  24. Reference processes Further insight on the quarkoniabehaviour in a hot matter can be obtained comparing the measured yield to a reference process not affected by the medium At SPS, J/ was studied wrtDrell-Yan process, not affected by the created medium …however low Drell-Yan rate at LHC. Several proposals: Z0, open charm, open beauty… Helmut’s golden reference best reference should share the same production mechanism with quarkonium initial and final state effects should be under control

  25. How can we compare to models? According to these rough estimates, the J/ centrality pattern could be affected by roughly: ~ 10% systematic +statistical error ~ 15% systematic error on J/pp(for RAA) ~18% total error however also large uncertainties on the size of relevant effects complicate the comparison if shadowing has ~50% systematic uncertainty, the experimental result, whatever its error, will hardly clarify any hot-matter related scenario 1 Meas/Exp (Ramona, priv.comm) A1 On top of this, experimental resolution should be taken into account, since it may wash out, for example, a step-wise suppression pattern A2 Npart Step position

  26. What about J/ from B? J/ from B can complicate even more the picture 7 TeVpp results show that the fraction of J/ from b hadrons does not strongly depend on energy  we can assume the same trend at 2.76 TeV Fraction of J/ coming from B do not suffer suppression in the medium CMS, arXiv:1011.4193 …but if B is strongly quenched by the medium, the fraction of J/ from b hadrons (vspT) will change More hints from ATLAS and CMS which will be able to separate prompt J/ from those from B decay also in PbPb (?)

  27. Comparison with lower s data Preliminary comparison of ATLAS and PHENIX data Centrality dependence of J/ suppression seems invariant with beam energy in spite of different • s (factor x14) • initial energy density (~3) • kinematic range (pT>0 for PHENIX, pT>6.5GeV ATLAS) • no B feed-down correction (4% PHENIX, 20% ATLAS) P. Steinberg, LPCC HI@LHC, March 2011 At a first glance…first LHC results do not seem to clarify all open questions… Maybe one should look also at open charm pattern (Helmut)

  28. What about the ? Bottomonium states should be a cleaner probe, accessible at LHC • More robust theoretical calculations • No b hadron feed-down …but with a lower production cross section Will we able to separate the  states? A good resolution is crucial to separate the 3  states, which have different dissociation temperatures . (1S) more easily separated (higher significance) Foreseen width: ~80-100 MeV S/B: ~0.3 - 2

  29. How many  can we collect? In the ALICE acceptanceroughly a factor 100 is expected between J/ and (1S) yields(J.Phys. G32,1295 (2006)) 2010 ~1500 J/  ~15 (1S) 2011~5000-8000 J/ ~50-80 (1S)  <20 (2S),(3S) ATLAS/CMS:  acceptance is larger than the J/ one, reaching pT~0  higher  statistics (ATLAS,J.Phys.G 34 (2007), CMS note 2006/89) 2010 ~700 J/  ~100 (1S) 2011 ~2500-4000 J/  ~400-600 (1S)  ~100-200 (2S)  <100 (3S) • centrality dependence out of reach in 2010? • …but maybe feasible in 2011!

  30. Conclusions Quarkonia study in heavy ion collisions is already a 25 years long story SPS and RHIC data provided many insight on the quarkoniumbehaviour in AA, but many questions are still open Many questions are looking for an answer from LHC data! the picture seems indeed quite complicate, …but, hopefully, putting together all the pieces of the puzzle the scenario will be clarified! CNM feed down SPS RHIC sequential melting regeneration J/ from B suppression LHC LHC

  31. Backup

More Related